Analysis of the Redundancy Parameterization of the ABB YuMi Robot Arm
The ABB YuMi robot arm, a 7-degree-of-freedom (DOF) collaborative robotic manipulator, presents distinct challenges in path planning due to its complex kinematic structure, absence of consecutive intersecting or parallel axes, redundant DOF, and restrictive joint limits. The authors address these challenges by thoroughly defining the redundancy parameterization through the Shoulder-Elbow-Wrist (SEW) angle, a crucial parameter needed for efficient manipulation and control. The paper provides the first detailed and validated definition of this SEW angle, aligning robotic behavior with observations made in the RobotStudio simulator, thereby offering a more complete kinematic descriptor for the YuMi arm.
In this study, the authors elucidate the ABB YuMi's intricate kinematic parameters, denoting the SEW angle as a parameterization for its redundancy. The SEW angle uses the direction of the fourth joint axis—aligned with conventional SEW angle formulation that considers shoulder, elbow, and wrist alignment—to navigate the arm's workspace efficiently. This detailed parameterization allows nuanced path planning beyond the capabilities of ABB's RobotStudio simulator's standard outputs, particularly in scenarios that necessitate working close to joint limits or in singularity-prone orientations.
The paper advances previous attempts to define the SEW angle that have been found either partial or incongruent with empirical data from RobotStudio. The authors supplement the SEW angle's definition with the computation of the SEW angle Jacobian and detailed conditions for algorithmic singularities, thus aiding in avoiding problematic robotic states which are not physically implementable by manipulations.
The strong numerical results underpinning this work include testing against multiple position states of the YuMi arm to affirm the accuracy of the SEW angle formulation. These outcomes are particularly noteworthy: the provision of complete inverse kinematic solutions with the use of IK-Geo, an inverse kinematics solver, confirms the practical applicability of the presented formulations. The authors demonstrate comprehensively how a 2D search can elucidate all inverse kinematic solutions, highlighting the robustness of their approach.
Furthermore, the paper explores both theoretical and practical implications of its findings. From a theoretical perspective, the paper advocates a meticulous framework for analyzing algorithmic and kinematic singularities, urging the development of future singularity-free redundancy parameterizations. Practically, a correct SEW angle parameterization enhances the arm's maneuverability in cluttered or constrained environments, which is critical for collaborative robots intended to safely interact with human partners.
Looking towards future implications, the paper touches upon potential expansions in algorithmic techniques to mitigate singularity risks further, such as implementing enhanced SEW angle models like the Stereographic SEW angle. Such refinements could streamline redundancy parameterization and mitigate singularities associated with specific manipulations.
In conclusion, this paper renders a comprehensive portrayal of the ABB YuMi's redundancy parameterization, providing invaluable insights into path planning in 7-DOF robotic systems. It establishes a foundation for further research and development within the domain, encouraging the refinement of redundancy parameterization and inspiring improvements in the control and application of modern robotic manipulators.
